The story is, the people who run the site made some aggregate statistics, and the statistics showed that women were getting much worse interview scores collectively.
They got curious as to whether their interviewers were showing some kind of bias against female applicants, so they ran a crude experiment. They used audio software to distort the voice of each candidate towards a male-sounding or female-sounding voice (the interviews are voice-only), creating four groups:
* Women distorted to sound more like men
* Women distorted to sound more like women
* Men distorted to sound more like men
* Men distorted to sound more like women
Then they conducted the interviews, and made some comparisons with a control group.
Long story short, they didn't uncover any systemic bias. They weren't willing to accept the face-value implication (that women are just worse at interviews) so they dug a little deeper and found something interesting:
Women were far more likely to get discouraged and quit the program after one or two bad interviews. Seven times more likely, in fact. 35% of all women who got a bad interview - that's one third - quit after that first bad interview, compared to 5% of men.
If the researchers removed these first-time and second-time quitters from the pool, the performance differences between men and women went away.
Now, a sample size of less than 300, on a web service that is subject to the whims of ad campaigns and selection-bias, is not definitive. But it brings up an interesting point to consider, and an even more interesting one after that.
* Is the problem here that women are too hard on themselves, relative to men?
* If so, is the solution to adapt the post-interview process so that rejections are delivered differently, to counteract the urge to quit?
I asked Kerry about this. She said she agreed with the first point. Early in her tech career she got one bad interview and it was devastating. She didn't apply for another job until a whole year had gone by, and she was much more confident in her skillset by then.
(I thought back to the beginning of my own career. The first real tech interview I remember was way back in 1993, at Atari. I was told, "you seem talented, but we're not going to hire a high-school student." The manager acted like an amused father trying to humor his over-ambitious son. It hadn't been a negative experience exactly - I knew it was a long-shot - but in retrospect, it was another three years before I actually applied for, and got, a job in tech.)
On the second point - that the post-interview process could be improved - Kerry agreed, but neither of us could figure out exactly how to change it.
But what about that first point again? Are women too hard on themselves generally? Do they tend to downplay their own talents and accomplishments, relative to men, who tend to brag and exaggerate? My gut tells me -- yes, absolutely.
Does this mean we need to make changes in the STEM universe on a broader scale, beyond the interview, in order to accommodate the more humble, deferential nature of women as a collective? Or does it mean that we should just accept this state of affairs at face-value, and let the women who can't hack it drop out into other careers, while the few that are happy in this male-dominated field stay the course? Is there actually anything we need to fix? Is the under-representation of women in this field just the result of personal preference? The cultural zeitgeist isn't calling for changes to the position of "sanitation worker" so that more women are encouraged to dump cans and clean sewers. Why is it calling for changes to the position of "software developer"?
I have some of my own answers to these navel-gazing questions. For one, I do not think the under-representation of women is due to personal preference, I think it is due to various conditions that are a legacy of the way the software industry started, making an environment that is arbitrarily hostile to women. The top two on my list are:
* A large base of eager male developers who were drawn into this career path by the gaming industry, which was heavily male-centric in marketing for decades after its inception, carrying along their own strange take on women. (Re: Lara Croft.)
* Too many brogrammers who have transitioned from college campuses directly to corporate campuses and carried along their behaviors, including hitting on women and being unprofessional with other men.
Also, I think the call to change the state of this field is legitimate, because the tech industry is a hugely important and growing one, and CEOs and hiring managers are clamoring to put more programmer asses in more seats, and keep them there. The average Joe doesn't have to care about the representation of women in tech -- the industry is what cares.
Also, there is another good reason to follow this thread. Men and women fall on a broad spectrum of temperament, and STEM fields are geeky fields where intellectual rigor and emotional perceptiveness need to go hand-in-hand. Men and women alike need to learn how to work harmoniously with people who can be a good standard deviation above the sensitivity level of the general public. We're not just talking about "apologize if you bump into them in the hall", we're on the level of "make sure you choose exactly the right words in your feedback to a comment on a pull request so you don't accidentally invoke a jihad over code formatting between your lead programmer and your project manager."
The (suspiciously) common wisdom is that men talk to prove themselves and gain dominance, and women talk to share and reach consensus. Well, I've been on a lot of teams over the years, and I can assure you that nothing is more refreshing than working with someone who admits their mistakes, owns them, and works to fix and prevent them, humbly recruiting others as needed. Is that more of a masculine trait, or a feminine one? I think the best answer is that it's a synthesis. And with too many bros strutting around, holding their egos out in front of them like squishy battering rams, that synthesis is hard to maintain.
One question to ask at this point is, how well does the ability to humbly negotiate consensus come across in the average job interview? Hah; I think it barely comes across at all. Young interviewers look for technical dexterity, since it's all they know how to judge. Older interviewers look for "fit", which can be subjective and capricious. Only if they're particularly wise, will they spend their allotted time with you judging your ability to negotiate conflict into consensus. ... But I can tell you, that is an archmage-level ability and if you find it, you hire it. With people like that you can build a team that punches out architecture like clockwork.
What this says to me is, the software industry - and perhaps all STEM fields - will function best by promoting a work environment that draws men and women towards a synthesis of their best traits. This is not a career like fire-fighting, where you need upper body strength, nor is it a career like early childhood education, where your experience as a caregiver in your own family gives you a leg up. This is a highly technical, highly articulate, highly cooperative pursuit. Neither women nor men have a monopoly here; we need to attract and retain both. And that means, we need to seek out and minimize the vestigial traits of it that are threatening to one or the other.
I went to Kerry's company's holiday party last year. They had women dressed as go-go dancers standing up on platforms along the walls, gyrating to the music. I had no idea how to interpret it, but it sure made me uncomfortable.
So yes, there are real changes worth making, and those changes are far from implemented. I haven't even mentioned the solid practical stuff, like on-site childcare, extended maternal leave, improved health insurance options, on-site charter schools, and outreach programs that mix work with recruitment and teaching efforts, for those (including myself) who would feel higher job satisfaction if they got to mentor as part of their career. Oh, and part-time or flexible month-on-month-off schedules, and better telecommuting integration.
But let's reel this back to the first major point that came up from the study:
If it is true that women in aggregate are discouraged "too easily" by negative feedback, then that presents a very real barrier.
In my last two interviews (both of which landed me jobs) I used the following line: "I believe three personality traits make me a good programmer. I am lazy, stubborn, and suspicious."
Then I waited a beat, and said (more or less) "I'm lazy so I'm constantly looking for a way to automate things. I'm stubborn, so when I hit a bizarre bug, I throw everything at it. And I'm suspicious, so I insert logging hooks, and write tests, and check configurations."
It gets a laugh and it's a good humble-brag at the same time. But the reason I bring it up is, you need stubbornness to do this job, because you have to smack your forehead against a wall most of the time you are at work. If you are not stuck, you are merely not caught up yet.
I was self-taught, and pushed ahead against an uncooperative and vexing machine to learn my trade, in a complete absence of positive feedback from any living soul. Sheer bloody-mindedness, as Sir Pratchett would call it, was at the very center of this pursuit and career from the beginning and to the present day.
This is really not a fulfilling career for anyone who is easily discouraged.
Now let me say, I have met many good female developers, and they all showed that same bloody-mindedness. I'm glad they're around, and I have respect for the added level of difficulty they face. (To borrow a famous quote, they are doing this job "backwards and in high-heels.") For example, it's a lot harder for them to communicate effectively when some indistinct group of male co-workers is keeping a minimum distance out of fear or spite, and another indistinct group is constantly distracting themselves by asking "is she flirting with me?" over and over in their heads every time they stand over the same screen to look at code.
Again, real changes do need to happen. But the point here is, are these women so few in number just because they are rare in the world to begin with? Is stubborn persistence in the face of rejection or failure, for year after year without burning out, just naturally less common in women? (Leaving aside the easily questioned meaning of "natural".)
If that is the case, then do we just have to accept that women will be under-represented in the software industry? What about other STEM disciplines?
My gut is telling me -- no. But unfortunately that's as far as I can take it, because I don't have numbers for the upcoming generation, that will hopefully take the reins of a tech industry that has undergone changes in their favor. Are more young women dedicating themselves to software programming? Are those numbers reaching up to parity with men? Will they actually make careers out of it? Or will they do it for a couple of years and then cast around for something else?
Actually, it doesn't matter, does it. It doesn't change our mission -- the point of these changes: To improve the efficiency and size of this industry as a whole, and to move closer to the ideal of judging every contributor by the quality of their work.